Benchmarking the cloud
Abstract
Our project aims to benchmark the various clouds for running scientific applications. This will usually test the quality of the network interconnect and the variation in the latency when running MPI jobs. Our secondary aim is to run any scientific applications which utilize transactional memory to test STM performance on the cloud
Intellectual Merit
How important is the proposed activity to advancing knowledge and understanding within its own field or across different fields?
- This project attempts to give all researchers a survey of various clouds out there to choose for their experimentation
How well qualified is the proposer (individual or team) to conduct the project? (If appropriate, the reviewer will comment on the quality of prior work.)
- We are a team of graduate students having a course in both cloud computing and scientific computing.
To what extent does the proposed activity suggest and explore creative, original, or potentially transformative concepts?
- It is a survey of existing tools/infrastructure in the market.
How well conceived and organized is the proposed activity?
-We have already run many scientific applications as benchmarks on the Amazon EC2 compute clusters. We also plan on running these on Open Science Data Cloud, subject to approval
Broader Impact
How well does the activity advance discovery and understanding while promoting teaching, training, and learning?
-This project, at the very least, will help researchers get started with using the clouds for scientific applications.
How well does the proposed activity broaden the participation of underrepresented groups (e.g., gender, ethnicity, disability, geographic, etc.)?
-Neutral to all these, it will open up the information for everyone to access
Will the results be disseminated broadly to enhance scientific and technological understanding?
-Yes
What may be the benefits of the proposed activity to society?
-Better understanding of the cloud infrastructures
Use of FutureGrid
Use the resources to benchmark the grid
Scale Of Use
A few VMs of each type of machine for a maximum of total 10 days